Research map / Reinforcement learning / Visual and representation RL
Scalable Trajectory Generation for Whole-Body Mobile Manipulation
Yida Niu, Xinhai Chang, Xin Liu, Ziyuan Jiao, Yixin Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace CPU-based AKR planning with GPU-accelerated batched trajectory optimization | AKR planner · Trajectory generation throughput | episodes per hour: 60 -> 5,000 per hour (over 80x faster) | Improved |
| use a mobile base instead of a fixed base (ablation) | DP3 policy, single scene · Microwave door opening | success rate (%): 100 (fixed, <800 traj) vs about 70 (mobile, 3,200 traj) | Hurt |
| increase number of training scenes from 1 to 30 (ablation) | DP3 policy · Unseen-scene door opening | success rate (%) | Improved |
| increase per-scene trajectory density with 30 scenes (ablation) | DP3 policy · Seen and unseen scenes door opening | success rate (%): about 75% success at high density | Improved |
| add more demonstrations within a single scene (ablation) | DP3 policy · Unseen IK start states, same scene | success rate (%) | Mixed |
| train DP and ACT instead of DP3 on the same data (ablation) | AutoMoMa data · 30-scene door opening | success rate (%) | Improved |
Closest papers by meaning
- DexFlyWheel: A Scalable and Self-improving Data Generation Framework for Dexterous Manipulation (NeurIPS 2025 · similarity 0.88)
- AC-DiT: Adaptive Coordination Diffusion Transformer for Mobile Manipulation (NeurIPS 2025 · similarity 0.88)
- CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects (NeurIPS 2025 · similarity 0.88)
- DynScene: Scalable Generation of Dynamic Robotic Manipulation Scenes for Embodied AI (CVPR 2025 · similarity 0.88)
- Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets (ICLR 2025 · similarity 0.87)
- Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
